Human embryonic stem cells (hESCs) are a highly promising source of differentiated cells that may be used in the future for revolutionary treatments of diabetes, Parkinson's and other diseases. hESCs have been somewhat difficult to study, because conditions that support their growth in culture were farily primitive. We have developed a simple and efficient way to culture hESCs, which opens the possibility of alternate and powerful cell culture methodologies, which may speed the progress of these clinical applications. The ojectives of the application are to develop key new enabling technologies for hESCs. We have established a new and simple defined media system for culturing and differentiating hESCs. Our approach supported the massive expansion of undifferenitiated cells, the maintenance of hESCs in suspension and the routine minaturization of hESC culture for high throughtput sceening approaches. The specific aims are to investigate the advantages of maintaining undifferenitated hESCs in suspension, and directing differentiation to endodermal and other lineages in high-density suspension culture. Streamlined protocols will be adapted to a fully-controlable bioreactor system and optimised. Other aims are to screen known compounds or libaries of marine natural products in hESCs to identify novel activities that impact self-renewal, differentiation or apoptosis. In particular, we will screen for compunds that induce differenitation in hESCs by impacting Activin/nodal, GSKSp or PIS Kinase signaling. Our approaches are highly unique and have the potential to greatly simplify future cell therapy applications of hESC-derived transplantable cells.